Winemaking and Bioprocesses Strongly Shaped the Genetic Diversity of the Ubiquitous Yeast Torulaspora delbrueckii.

OPEN 3 ACCESS Freely available online

•0-PLOS I ONE

Winemaking and Bioprocesses Strongly Shaped the (S\
Genetic Diversity of the Ubiquitous Yeast Torulaspora
delbrueckii

Warren Albertin 1,2 *', Laura Chasseriaud 1,2 ', Guillaume Comte 1 , Aurelie Panfili 1 , Adline Delcamp 3 ,
Franck Salin 3 , Philippe Marullo 1 ' 2 , Marina Bely 1

1 Univ. de Bordeaux, ISVV, EA 4577, Unite de recherche CEnologie, Villenave d'Ornon, France, 2 Biolaffort, Bordeaux, France, 3 INRA, UMR Biodiversite Genes et
Ecosystemes, PlateForme Genomique, Cestas, France

Abstract

The yeast Torulaspora delbrueckii is associated with several human activities including oenology, bakery, distillery, dairy
industry, etc. In addition to its biotechnological applications, T. delbrueckii is frequently isolated in natural environments
(plant, soil, insect). T. delbrueckii is thus a remarkable ubiquitous yeast species with both wild and anthropic habitats, and
appears to be a perfect yeast model to search for evidence of human domestication. For that purpose, we developed eight
microsatellite markers that were used for the genotyping of 110 strains from various substrates and geographical origins.
Microsatellite analysis showed four genetic clusters: two groups contained most nature strains from Old World and
Americas respectively, and two clusters were associated with winemaking and other bioprocesses. Analysis of molecular
variance (AMOVA) confirmed that human activities significantly shaped the genetic variability of T. delbrueckii species.
Natural isolates are differentiated on the basis of geographical localisation, as expected for wild population. The
domestication of T. delbrueckii probably dates back to the Roman Empire for winemaking (~1900 years ago), and to the
Neolithic era for bioprocesses (~4000 years ago). Microsatellite analysis also provided valuable data regarding the life-cycle
of the species, suggesting a mostly diploid homothallic life. In addition to population genetics and ecological studies, the
microsatellite tool will be particularly useful for further biotechnological development of T. delbrueckii strains for
winemaking and other bioprocesses.

Citation: Albertin W, Chasseriaud L, Comte G, Panfili A, Delcamp A, et al. (2014) Winemaking and Bioprocesses Strongly Shaped the Genetic Diversity of the
Ubiquitous Yeast Torulaspora delbrueckii. PLoS ONE 9(4): e94246. doi:10.1371/journal.pone.0094246

Editor: Joseph Schacherer, University of Strasbourg, France

Received November 19, 2013; Accepted March 14, 2014; Published April 9, 2014

Copyright: © 2014 Albertin et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits
unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.

Funding: This research was funded by the Univ. de Bordeaux. This work was supported, in part, by the European commission in the framework of the FP7-SME
project Wildwine (grant agreement n 315065). The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the
manuscript.

Competing Interests: The co-authors Laura Chasseriaud and Philippe Marullo are affiliated with BIOLAFFORT. This does not alter the authors' adherence to all
the PLOS ONE policies on sharing data and materials.

* E-mail: warren.albertin@u-bordeaux.fr

9 These authors contributed equally to this work.

Introduction

The remarkable physiological properties of yeasts have led to
their wide use in the field of biotechnology. Their fermentative
ability has been exploited by humans for millennia to ferment and
preserve beverages and food. The most famous yeast is undoubt-
edly Saccharomyces cerevisiae, which has been making bread, beer [1],
wine [2] and spirits [3] -the oldest applications- for ages, and more
recently bioethanol [4] . Indeed, population genetics of S. cerevisiae
reveals a strong connection to human civilization and history, with
genetic clustering coinciding with biotechnological applications
[5] . This indicates that over hundreds or even thousands of years,
human uses promoted the adaptation of S. cerevisiae to various
food/beverage systems, a process called domestication [6].

Besides Saccharomyces cerevisiae, several dozen yeast species are
involved in various biotechnological processes, such as Ogataea
angusta (formerly Hansenula polymorpha) for production of recombi-
nant proteins among which pharmaceuticals [7], Komagataella
(Pichia) pastoris for production of pharmaceutical/nutrient com-
pounds [8], Kluyveromyces lactis var. lactis for enzymatic production

[9] . However, there are far less data regarding population studies
of so-called "non-conventional" yeasts. Indeed, while there are
several examples of adaptation of molds and lactic acid bacteria to
anthropic food environments [6], it is still unclear to what extent
the domestication process shaped yeast evolution.

In this work, we considered a non-conventional yeast species of
technological interest, Torulaspora delbrueckii. T. delbrueckii has been
associated with winemaking for decades [10-12] and isolated
either from grape, must or wine. Although T. delbrueckii is generally
unable to complete alcoholic fermentation [i.e. to consume all
sugars), it produces relatively high ethanol concentrations for a
non- Saccharomyces yeast [11,13-15]. This explains why T. delbrueckii
was formerly classified within the .Saccharomyces genus (under S. rosei
or S. roseus name). T. delbrueckii also produces low levels of
undesirable volatile compounds (hydrogen sulphide, volatile
phenols) [16,17], reduces volatile acidity in high-sugar fermenta-
tions when associated with S. cerevisiae in mixed cultures [17] and
increases sensorial complexity [18-20]. Due to its oenological
interest, the duo formed by S. cerevisiae and T. delbrueckii is now

PLOS ONE | www.plosone.org

1

April 2014 | Volume 9 | Issue 4 | e94246

Microsatellite Analysis of T. delbrueckii

becoming a model for studying interaction mechanisms between
yeast populations [21-23].

Besides its potential for winemaking, T. delbrueckii has biotech-
nological applications in the bread industry [24] due to dough
leavening ability associated with high freezing and osmotic
tolerance [25-27]. T. delbrueckii is frequently described as a major
component of yeast biota from dough carried over from previous
bread making in rural regions [24] or from artisanal bakeries [28].
Commercial exploitation of this species has recendy begun, with
some T. delbrueckii strains that are commercialized in Japan for
frozen dough applications [24].

T. delbrueckii is also naturally associated with several other
human bioprocesses, ranging from food fermentations of silage,
cocoa [29,30], olive [31] or cucumber [32,33], to distilled and
traditional fermented beverage production including mezcal [34],
colonche [35], tequila [36], cider [37], strawberry tree fruits juice
[38], sugarcane juice [39,40] and kefir [41]. T. delbrueckii is a
frequent component of dairy products' microflora, either as
desirable ferment for traditional cheeses [42] and fermented milk
[43], or as spoilage yeast [44,45]. Other processed products like
soft drinks (fruit juices, etc.) can be spoiled by yeasts including T.
delbrueckii [46].

In most bioprocesses where T. delbrueckii is identified, the species
is not added deliberately, unlike S. cerevisiae which is usually added
as lyophilized inoculum. 7 '.delbrueckii naturally colonized a wide
range of anthropized habitats. This species is also frequendy
isolated from natural environments, ranging from soils [47], to
plants [48], fruits [49] and insects [50,51]. Finally, this species,
although not considered to be a human pathogen, is occasionally
found as a clinical isolate [52] where it is usually referred as
Candida colliculosa, the anamorphic form of T. delbrueckii [53].

Thus, Torulaspora delbrueckii is a remarkably ubiquitous yeast
species with both natural reservoirs and habitats associated with
human activities (winemaking and other bioprocesses), and
appears to be a perfect yeast model to search for evidence of
domestication besides the baker's yeast S. cerevisiae. However, to
date, few tools for molecular characterization of T. delbrueckii
strains were available. Mitochondrial RFLP appears too poorly
discriminant at the strain level [24]. Restriction endonuclease
analysis associated with pulse-field gel electrophoresis (REA-
PFGE), although discriminant [16], is time-consuming and does
not allow for accurate population genetics or ecological studies. In
this work, we developed eight microsatellite markers for the T.
delbrueckii species. This new tool was used for the genotyping of 1 10
strains from various geographical regions and various substrates.
We show that the genetic variability of the ubiquitous yeast T.
delbrueckii is strongly shaped by human activities and in particular
by winemaking and other bioprocesses.

Materials and Methods

Yeast strains and culture conditions

One hundred and ten strains of Torulaspora delbrueckii were
sampled from various isolation substrates (grape/wine, nature,
clinical, bakery, spoiled food, fermented beverages, dairy products
and other bioprocesses) and from worldwide locations (Figure 1,
Table SI and Figure SI). In addition, the type strains of T.
franciscae, T. pretoriensis, T. microellipsoides, T. globosa, T. indica [54], T.
maleeae [55], and T. quercuum [56] were used to test whether the
microsatellites developed for T. delbrueckii could be useful for other
Torulaspora species.

All strains were grown at 24°C in YPD-based medium
containing 1% yeast extract (w/v, Difco Laboratories, Detroit,
MI), 1% Bacto peptone (w/v, Difco), and 6% glucose (w/v),

supplemented or not with 2% agar (w/v). For a quick assessment
of respiratory-ability, cells were plated on YPGly medium,
containing glycerol as unique source of carbon (1% yeast extract
(w/v, Difco), 1% Bacto peptone (w/v, Difco), 2% (v/v) glycerol
and 2% (w/v) agar). A minimum medium SD containing 0.67%
Yeast Nitrogen Base (w/v, Difco), 2% glucose (w/v) and 2% agar
(w/ v) was used to test for prototrophy/auxotrophy. The ability to
sporulate was checked by microscopy after 3 days at 24°C on
acetate medium (1% potassium acetate, 2% agar). T. delbrueckii
strains usually formed asci with one unique ascospore, while ascii
with two to four spores were more rare.

Genomic DNA extraction

For DNA extraction, cells grown on YPD medium were lysed
using a FastPrep-24 instrument (MP Biomedicals, Illkirch, France):
100 uL of glass beads (acid-washed, 425-600 urn, Sigma, Lyon,
France) were added to cells pellet as well as 300 ul od Nuclei Lysis
solution (Wizard Genomic DNA purification Kit, Promega). Cells
were crushed through 2 cycles of 20 s (max. speed). Subsequent
DNA extraction was performed with the Wizard Genomic DNA
purification Kit (Promega) following the manufacturer's protocol.

Species assessment

PCR-RFLP of the ITS region (with EcoSA digestion) was
performed as described by Granchi et al. [57] to confirm the
Torulaspora genus. In addition, we developed two additional PCR-
RFLP markers to discriminate T. delbrueckii strains from the other
Torulaspora species: briefly, we amplified D1/D2 domain by means
of universal primers NL1 and NL4 [58].

Enzymatic digestions of the 600 pb amplicon were carried out
on 10 ul of amplified DNA in a final volume of 1 5 ul with either
AM or Pstl (New England Biolabs, Ipswich, MA) for 16 h at 37 °C.
Restriction fragments were separated by a microchip electropho-
resis system (MultiNA, Shimadzu). Alul restriction allowed
discriminating T. delbrueckii from all other species except T.
quercuum, while Pstl digestion showed different restriction patterns
for T. delbrueckii and T. quercuum (Figure 2).

Microsatellite loci identification and primer design

Dinucleotide to tetranucleotide repeats were identified within
the genome sequence of T. delbrueckii type strain CBS 1 146 T (CLIB
230 T ) [59]. In order to exclude possible telomeric and subtelo-
meric repeats, we did not consider microsatellites located within
3Kb of the 5'-end or 3'-end of the contigs. Primers were designed
using the 'Design primers' tool on the SGD website (http:/ /www.
yeastgenome.org/cgi-bin/web-primer). In order to reduce the cost
associated with primer fluorescent labelling, the forward primers
were tailed on the 5'-end with the M13 sequence (19 nt) as
described by Schuelke, 2000 [60], allowing the use of M13 primers
labelled with FAM or HEX for different PCR reactions. Amplified
fragment sizes varied from ~ 1 10 to ~380 bp, allowing subsequent
multiplexing of the amplicons (Table 1).

Microsatellite amplification

PCR reactions were performed in a final volume of 10 ul
containing 50-100 ng of genomic DNA, 0.05 uM of forward
primer, 0.5 uM of reverse primer and labelled primer, IX Taq-
&GO (MP Biomedicals, Illkirch, France). Universal primers Ml 3
were labelled with either 6-carboxyfluorescein (FAM), or hexa-
chlorofluorescein (HEX) (Eurofms MWG Operon, Les Ulis,
France).

PCR was carried out using a thermal cycler (iCycler, Biorad,
Hercules, CA, USA) as followed: Initial denaturation step (5 min

PLOS ONE | www.plosone.org

2

April 2014 | Volume 9 | Issue 4 | e94246

Microsatellite Analysis of T. delbrueckii

Origin

• bioprocess

^ clinical

■ grape/wine

o nature

▼ N/A

Figure 1. Geographical localisation of the T. delbrueckii strains used in this study. 3 strains (CUB 230, CUB 503, MUCL 27828) are not
represented since their precise isolation location is unknown. More details on European isolates (grey box) can be found on Figure SI.
doi:1 0.1 371 /journal.pone.0094246.g001

at 95°C) followed by 35 cycles of 35 s at 95°C, 50 s at melting
temperature (see Tm in Table 1) and 40 s at 72°C, and a final
extension step of 7 min at 72°C.

Amplicons were initially analysed by a microchip electropho-
resis system (MultiNA, Shimadzu) and the optimal conditions for
PCR amplifications were assessed. Then, the sizes of the amplified
fragments were measured on an ABI3730 DNA analyzer (Applied
Biosystems). For that purpose, PCR amplicons were diluted (1800-
fold for FAM and 600-fold for HEX-labelled amplicons respec-
tively) and multiplexed (Table 1) in formamide. LIZ 600 molecular

marker (ABI GeneScan 600 LIZ Size Standard, Applied
Biosystem) was 100-fold diluted and added for each multiplex.
Before loading, diluted amplicons were heated 4 min at 94°C.
Allele size was recorded using GeneMarker Demo software V2.4.0
(SoftGenetics).

Data analysis

Microsatellite analysis was used to investigate the genetic
relationships between strains. A dendrogram was built using
Bruvo's distance and Neighbor-Joining clustering, by means of the

bp

500
350
250
200
150
100
75
50

25

58

;Q CM

X -=t

3 °>

-S? o

CD *"

c co

c CD

Ico

3 O

-2

CO

o

CO ^-

.a cm

"O »-

C CO
CO

a.

h

g in

S in

CD CM

Co

So
.a m
o <°

"a?

__l LL1 UJ LL1 |

—

Figure 2. Restriction patterns of D1/D2 amplicon generated by AliA (A) or Pst\ (B) for Torulaspora species. A: For Alu\ restriction, four
patterns were produced: 170 pb + 160 pb + 80 pb + 70 pb + 55 pb+40 pb + 30 pb for T. delbrueckii and T. quercuum;
170 pb+160 pb+120 pb+70 pb+55 pb+30 pb for T. maleeae and T. indica; 170 pb+160 pb+95 pb+80 pb+70 pb+30 pb for T. franciscae, T.
microellipsoides, T. pretoriensis; and 330 pb+170 pb+75 pb+30 pb for T. globosa. B: For Psfl restriction, two patterns were produced: 600pb (no
restriction) for T. maleeae, T. quercuum, T. indica, T. microellipsoides and T. globosa; or 480 pb+1 20 pb for T. delbrueckii, T. franciscae and T. pretoriensis.
Blue and pink bands represent internal upper and lower markers respectively.
doi:10.1371/journal.pone.0094246.g002

PLOS ONE | www.plosone.org

3

April 2014 | Volume 9 | Issue 4 | e94246

Microsatellite Analysis of T. delbrueckii

<V * -Q IN

« tj E £

^ 01

iii a. E
< i c

hi

.0 S

£ =

1 2

8 g

II

U a

Q.

>s

O
C
0)
Ol

■Q

s

o
a

a
"3

IS

o

o

CM

ro

O

fN

vO

r-»

CO

CO

Oi

r--
ro

fN

v£>

CO

rsi

r\i
Ln
fN

m

'aj
o

a o

Id °

u ro

'■P O

<U <

£ 3

O Qj

S-a

<M So
>— m
10 IN

<

5y

St

U <3

Sb

u <

< <

< <

U <3

< <

13 13

< U
„ 13

ai .E

— . <u

F 2
=i _ \D

, too

4-< O

IS

cn

ro

ON

ro
O

(4:25)

fN
ro

(1:1

LO
CM

ro

£ § E

O vo O in

E °

E 2
o ^-

E 2 E

o m o o o o

ID
13
13

i?

p

h- U

< t

< ^

< U

< U
h- U
13 h-
„ U

<

in <

y y

3 y

t5 E

13 F

h- U

< <
13 <

< <
13 u

!; <

p

13

r

U 13

13

13 <

U <

U <

< <c

„ 13

u- U

yp

gi

13 C

13 ^

13 F

< U

< 13
U U

< U

< U
„ U
u- <

13

2 <

F

g y

13 F

13 13

< 13

13 H

13 <3

g s

<

a. o

SrP

5?

s °

c vo cm
O ro O ro

o o o cri om om o3; ocr>

o vo o m

u

F H

13 u

13 <

cy
y 3

< !3
U !3
F U

< u

< <

<3

<

13
<

<

= CQ O
< U

PLOS ONE | www.plosone.org

4

April 2014 | Volume 9 | Issue 4 | e94246

Microsatellite Analysis of T. delbrueckii

R program [61] and the following packages: polysat vl.3 [62], ape
[63], adephylo [64], phyclust [65]. Bruvo's distance takes into
account the peculiar mutational process of microsatellite loci, and
is particularly well adapted for populations with mixed ploidy
levels [66]. In order to assess the robustness of tree nodes,
multiscale bootstrap resampling associated with an approximately
unbiased test [67] was performed by means of R and the pvclust
package vl.2-2 [61,68].

In addition to dendrogram drawing, the software structure
(v2.3.4) was used to delineate clusters of individuals on the basis of
their microsatellite genotypes using a Bayesian approach [69].
Only strains with a maximum of two alleles per loci (i.e. considered
as diploids) were conserved for structure analysis (104 strains
upon 110). The parameters were as followed: 10000 Burn-in
period, 1000 Repetitions. Models with number of populations (K)
ranging from K— 3 to K= 20 were tested, and models with and
without admixture gave similar results (the model with no
admixture was thus conserved for the graphical representation of
the population).

To test for population differentiation, analysis of molecular
variance (AMOVA) was performed by means of the pegas package
[70] with n= 1000 permutations. We tested whether the genetic
distance was significandy explained by substrate origin and by
geographical localisation (i.e. the continent of isolation was used as
grouping factor). F-Statistics (F ST , F IT , F IS ) were computed for each
locus using Weir and Cockerham formula [71], and we tested
whether the genotype frequencies of each locus followed the
Hardy— Weinberg equilibrium by means of the pegas package [70].

In order to obtain an estimate of the divergence time between
different T. delbrueckii genetic clusters from the microsatellite data,
we used the method described by Goldstein et al. [72]. In yeasts,
the mutation rate for microsatellites generally falls between 1.10 4
and 1.10 6 per cell division [73,74], so we used an average
mutation rate of 1 . 1 0 5 per cell division. The number of
generations per year in yeast populations is difficult to estimate
correctly: Fay and Benavides considered a maximum of 2920
generations per year for domesticated S. cerevisiae [75], but this may
be considerably lower for wild yeast populations [76] for which we
considered a 100 generations per year as a maximum.

Results

Development of polymorphic microsatellite markers for
Torulaspora delbrueckii

We took advantage of the recent de novo assembly of the genome
sequence of CBS 1 146 1 (synonymous to CLIB 230 1 ) [77], the type
strain of Torulaspora delbrueckii, to search for microsatellite loci. We
only considered dinucleotide to tetranucleotide repeats that were
not located within the 5 '-end and 3 '-end of the chromosomes, in
order to exclude possible telomeric or subtelomeric positions. We
retained eight microsatellite loci that were located on six of the
eight chromosomes of CBS 1146 1 (Table 1). Some loci were
located in coding regions, like TD5A (low-affinity hexose
transporter LGT1 gene) or TD1A, TD2A, TD6A and TD8A,
located in hypothetical protein coding sequences. The three
remaining loci, TD1B, TD1C and TD7A were in non-coding
regions.

The amplicons were separated using a microchip electropho-
resis system (MultiNA), and the optimal conditions for microsat-
ellites amplifications were assessed on a panel of twenty strains of
T. delbrueckii (data not shown). After optimisation on T. delbrueckii
strains, the microsatellites markers were tested on seven additional
species of the Torulaspora genus: T. franciscae, T. pretoriensis, T.
microellipsoides, T. globosa, T. indica [54], T. maleeae [55], and T.

quercuum [56]. All these species appear to produce good
amplification of several markers, suggesting that some of the
microsatellites developed for T. delbrueckii could be useful for the
study of population genetics of other Torulaspora species. In
particular, the type strain of T. pretoriensis Y-17251 T showed a
complex genotype, with three alleles for three loci (TD1B, TD2A,
TD6A), two alleles for three loci (TD1A, TD5A, TD8A) and one
allele for the remaining two loci (TD1C, TD7A). This suggested T.
pretoriensis Y-17251 T could be aneuploid or polyploid. Further
characterization of additional T. pretoriensis strains will help
determine whether all strains share complex genome or whether
complex genotype is strain-specific.

The eight microsatellites markers were then used to genotype
1 1 0 T. delbrueckii strains isolated from worldwide regions (Figure 1 ,
Figure SI) and from various substrates (Table SI): 34 strains were
natural isolates isolated from plants, insects, soil, etc., 36 strains
were associated with several bioprocesses including bakery, cider
brewery, dairy processes, other fermented beverage and food
industries, excluding winemaking. While winemaking is a biopro-
cess, a particular focus was placed on strains from grape/wine
habitats due to long historical association between winemaking
and T. delbrueckii, in comparison to other bioprocesses. Thus, 35
strains related to winemaking and isolated from grapes, must, wine
or oenological material, were included. Finally, 3 clinical strains
were also included in the collection, as well as 2 strains of unknown
origin (one of which was the type strain CBS 1146 T = CLIB
230 1 ). All strains were able to grow on minimal medium (SD) and
YPG medium containing glycerol as the sole source of carbon,
indicating that all 110 strains were prototrophs and able to respire.
In addition, all 110 strains studied here were able to sporulate and
usually form ascii with one to four ascospores, ascii with a single
ascospore were far more frequent than those with 2 to 4
ascospores.

Genotyping the 110 strains of our collection revealed that all
microsatellites were polymorphic, with 6 different alleles for TD7A
and up to 40 alleles for TD8A (Table 1, Table S2). For some loci,
some strains showed allele size incompatible with the stepwise
mutation model predicting the increase or decrease in repetition
number and thus strict motif multiplication [78]. This indicated
that some punctual insertion/ deletion may arise either within the
microsatellite locus itself, thus increasing motif complexity as
previously shown [79,80], or within adjacent amplified sequence.
Finally, all 110 strains tested showed unique genotype, confirming
the discriminant power of microsatellite analysis.

Human activities shaped the genetic variability of
Torulaspora delbrueckii species

The genetic relationships between the 1 10 strains of T. delbrueckii
were further examined using Bruvo's distance (which is particu-
larly well-adapted for microsatellite data and populations with
unknown/ variable ploidy levels) and Neighbor-Joining clustering.
The resulting dendrogram tree showed four main clusters that
were strongly related to substrate origin (Figure 3A). Two groups
contained most strains isolated from nature, with a clear
dichotomy depending on their geographical origin. Indeed,
"nature/Americas" group comprised 25 strains of which 12 were
isolated from plants, insects or soils, and 16 isolated on the
American continent (with representatives from North, Central and
South America). The nature /Americas group was moderately
supported with bootstrap value of 51, due to the uncertain position
of CBS6518 strain. The descending node (excluding CBS6518)
was much more robust (bootstrap value of 89). The "nature/Old
World" group comprised 24 strains, 12 out of 24 were indeed
isolated from nature (plants and soils), and 18 out of 24 were

PLOS ONE | www.plosone.org

5

April 2014 | Volume 9 | Issue 4 | e94246

Microsatellite Analysis of T. delbrueckii

nature nature bioprocess grape/wine

Figure 3. Genetic relationships between 110 T. delbrueckii strains using eight microsatellite markers. A: Dendrogram tree built using
Bruvo's distance and Neighbor-Joining's clustering. The robustness of the node was assessed using multiscale bootstrap resampling and
approximated unbiased test (n = 1000 boots). Bootstrap results are shown only for the main nodes. B: Barplot representing structure results (K = 5).
The posterior probability (y-axis) of assignment of each strain (vertical bar) to ancestral groups is shown by colors (dark green, green, blue, red and
darkblue colors represent each 5 ancestral populations). Heterozygous strains, meaning strains with at least one heterozygote locus, are indicated by
black stars.

doi:10.1371/journal.pone.0094246.g003

isolated from the Old World (Europe, Asia and Africa). Nature/
Old World group was very robust (bootstrap values of 94).

The third cluster, designed as "grape/wine" group, was
composed of 27 strains, most of them (21/27) being isolated from
grapes or wines. The grape/wine group was moderately supported
(bootstrap value of 58), due to uncertain position of H strain.
However, inferior node (excluding H strain) was much more
robust (79). Interestingly, this group was not structured according
to the geographical origin, and the main wine regions of the world
were all represented (several European regions, California and
South America, Australia and New Zealand). The last group was
moderately supported (bootstrap value of 55) and contained 34
strains, of which 18 were associated with various bioprocesses and
human activities (excluding winemaking) from the five continents.
Noticeably, within this so-called "bioprocess" group, a sub-cluster

containing mosdy dairy strains was observed (6 of 1 1 strains), with
a moderately supported bootstrap value (55). Analysis with the
program structure was congruent with the dendrogram tree:
structure found an optimum of K— 5 populations that captured
the major genetic structure of T. delbrueckii species (Figure 3B).
These populations were consistent with the four genetic clusters
previously defined from the dendrogram tree (nature/Americas,
nature/Old World, grape/wine and other bioprocesses), and also
supported the dairy group (Figure 3B).

In order to definitively determine whether, and to what extent,
the genetic variation of T. delbrueckii was related to substrate and/
or geographical origin, an analysis of molecular variance
(AMOVA) was performed. We used either the substrate origin
(nature, bioprocess, clinical, grape/wine), or the geographical
localisation (using continent of isolation) as grouping factors. The

PLOS ONE | www.plosone.org

6

April 2014 | Volume 9 | Issue 4 | e94246

Microsatellite Analysis of T. delbrueckii

substrate origin explained 12.29% of the total variation of
microsatellite dataset, and was strongly significant (p-val-
ue<0. 00001), indicating that substrate origin shaped significantly,
yet not completely, the T. delbrueckii population structure. By
contrast, the geographical localisation, although significantly
related to molecular variation (p-value<0.001), explained less
variation (7.61%). It should be noted that the geographical
localisation, when considering only strains from nature, explained
17.19% of the genetic variation of nature isolates (and was
significant with p-value<0.001), while it was no longer significant
when considering strains from bioprocess and grape/wine origins.
This confirmed that human activities, namely winemaking or
other bioprocesses, significantly shaped the genetic variability of
the corresponding T. delbrueckii strains, while nature isolates are
differentiated on the basis of geographical localisation, as expected
for a wild population.

Torulaspora delbrueckii is a highly inbred species

Individually, the 1 1 0 T. delbrueckii strains included in the analysis
had one to 4 alleles per locus. Most strains (71 out of 110) were
homozygous for all eight loci, while 33 out of 110 presented a
maximum of two alleles per locus. Five strains (B 1 7 2 , CBS2924,
DC2, DIL 1 13 and UWOPS 79-138) had a maximum of 3 alleles,
but for one locus only, the hallmark of punctual genetic
duplication or aneuploidy rather than whole genome duplication
(polyploidy). Finally, UWOPS 83-777.1 had 3 and 4 alleles for
TD1A and TD5A respectively, suggesting a more complex
genome structure.

Considering T. delbrueckii to be a diploid species, we calculated
different population parameters (Table 2). Observed heterozygos-
ity (Ho) was low, from 0.009 for TD7A (the less polymorphic
microsatellite locus) to 0.128 for TD1C and TD5A (mean
Ho = 0.087). Indeed, the population data strongly deviated from
Hardy-Weinberg expectations for all 8 microsatellite loci, with
excess in homozygous strains (Table 2). This was particularly
evident for the "grape/wine" group, with only three out of 27
heterozygous strains (Figure 3). To determine whether such high
level of homozygosity was due to inbreeding and/ or to subpop-
ulation differentiation, we calculated F-statistics (F ST , F IS and F JT>
Table 2). All 8 loci gave similar results: a high total deficit of
heterozygotes {F r/ j, i-t. close to 1, associated with high deficit of
heterozygotes within the population (F/s), indicating that the excess
of homozygote individuals was mainly due to high inbreeding

Table 2. F-statistics and observed heterozygosity in
Torulaspora delbrueckii population.

Microsatellite

F,r

F ST

F,s

Ho

TD1A

0.914

0.034

0.911

0.103 ***

TD1B

0.906

0.104

0.896

0.120 ***

TD1C

0.877

0.020

0.874

0.128 ***

TD2A

0.945

0.118

0.938

0.077 ***

TD5A

0.861

0.162

0.834

0.128 ***

TD6A

0.972

0.140

0.968

0.034 ***

TD7A

1.00

0.065

1.00

0.009 ***

TD8A

0.921

0.050

0.917

0.094 ***

Fu represents the total deficit of heterozygotes, F ls the deficit of heterozygotes
within the population, F 5 t the fixation index. *** indicates a significant effect at
0.1%. Ho stands for observed heterozygosity, and did not fit the Hardy-
Weinberg hypothesis (pval<<0.001) for all eight loci.
doi:1 0.1 371 /journal.pone.0094246.t002

within each subpopulation rather than to subpopulation differen-
tiation.

Estimating the divergence time of Torulaspora delbrueckii
genetic groups

We estimated the divergence time between the different T.
delbrueckii genetic clusters from microsatellite data using the
method described by Goldstein et al. [72]. We used an average
mutation rate of 1 . 1 0 per cell division, as the mutation rate for
microsatellite in yeast generally falls between 1.10 4 and 1.10 6
per cell division [73,74]. The number of generations per year in
wild yeast populations is difficult to estimate correctiy [76], but we
considered a 100 generations per year as a maximum. Indeed, in
winemaking and most bioprocesses, the population size of
indigenous T. delbrueckii completed 7 generations per process [81],
which can be repeated a few times per year, so that 1 00 generations
per year seems to be a maximum. For wild strains, knowing that
yeast growth requires a combination of favourable physico-
chemical conditions as well as nutrient availability (sugar,
nitrogen), we assumed wild strains would only grow during
spring/ summer which may limit their growth to less than one
hundred generations a year. Using these parameters, the grape/
wine cluster was estimated to diverge from the nature/ Old World
group 1908 years ago [95% interval confidence: 1233-2125 years
ago]. By contrast, the bioprocess group was older and diverged
from the nature/ Americas cluster 3882 years ago [95% interval
confidence: 2961-5671 years ago].

Discussion

Torulaspora delbrueckii is a domesticated species for
winemaking and bioprocesses

Microsatellite genotyping is widespread for population, ecolog-
ical and evolutionary studies of eukaryote species [82,83], and
provided new insights into the population structure of S. cerevisiae
yeast [5,84-86]. In particular, the strong relationship between
genetic clustering and biotechnological applications [5] indicated
that S. cerevisiae was a domesticated species. The present study
shows that T. delbrueckii strains also cluster depending on their
human use, with the existence of genetic groups connected to
winemaking and bioprocesses. Many groups had high bootstrap
values, the likeliness of the four groups was further confirmed by
structure analysis, and the relationship between genetic variation
and substrate origin was statistically validated by AMOVA.

The "grape/wine" group is particularly interesting, with strains
from various continents demonstrating more genetic proximity
than strains from the same continent but from different substrates
(bioprocess, nature). This suggests that a group of closely related
individuals gave rise to the "grape/wine" population, and were
selected, consciously or unconsciously, for wine production. For S.
cerevisiae, the wine domestication event occurred 10 000-12 000
years ago (coinciding with the first archaeological records of
winemaking), indicating that S. cerevisiae was selected at the very
beginning of wine production [5]. To determine if the wine
domestication of T. delbrueckii preceded that of S. cerevisiae, the
divergence time between the different T. delbrueckii genetic clusters
was estimated from microsatellite data. The grape/wine cluster
diverged from nature/Old World group around 1900 years ago,
suggesting that T. delbrueckii domestication for winemaking is much
more recent than S. cerevisiae and is related to the modern history of
oenology. More precisely, T. delbrueckii domestication is contem-
porary with the Roman Empire where Vitis vinifera expanded
throughout Europe [87]. In the Middle Ages, V. vinifera further

PLOS ONE | www.plosone.org

7

April 2014 | Volume 9 | Issue 4 | e94246

Microsatellite Analysis of T. delbrueckii

expanded throughout the Old World with religions extension, i.e.
Christian crusades in Northern Europe, Islam spread to North
Africa and Middle East [87]. As T. delbrueckii is frequently isolated
on grapes and other plants, we hypothesized that an ancestral
"grape/wine" population of T. delbrueckii spread all around the
world with grapevines varieties during the Roman Empire and the
Middle Ages, and that their progeny was thus associated with
vinification practices in the different wine regions.

Remarkably, besides its use for winemaking, T. delbrueckii strains
are associated with several other human activities (dairy products,
bakery, distillery, other food and beverage fermentation) that
clustered together. The bioprocess group was estimated to be older
than T.delbrueckii domestication for wine and to date back around
four millennia ago, suggesting simultaneous anthropization of T.
delbrueckii and S. cerevisiae for several food and beverage processes
during the Neolithic era [88]. Within the bioprocess group, a sub-
specialization was observable for dairy products, like S. cerevisiae for
which specialization for beer, bread or sake was reported [75,88].
In addition to the grape /wine and bioprocess clusters, we
identified two groups mosdy containing strains from nature,
indicating that T. delbrueckii consists of both wild and domesticated
populations, as shown for S. cerevisiae [75].

Although microsatellite analysis showed strong evidence for
wine and bioprocess domestication, the genetic clustering was not
perfect, suggesting frequent exchanges between subpopulations
that may be mediated by insects or human activities as suggested
for S. cerevisiae [86,89]. To date, T. delbrueckii is the only non-
Saccharomyces yeast proved to be domesticated and thus
constitutes a complementary model system to the Saccharomyces
genus. Population genetics of other yeasts of biotechnological
interest will help determine whether anthropization shaped
significandy the genetic variability of various yeasts besides S.
cerevisiae and T. delbrueckii.

Understanding the life-cycle of Torulaspora delbrueckii

In addition to population structure, microsatellite analysis may
be useful for understanding the life-cycle [90] . Different life cycles
could be congruent with the results presented here, the main ones
being either a mostly diploid life-cycle, with both homozygous and
heterozygous homothallic diploid representatives, and a few
aneuploid/polyploid individuals; or a mostly haploid life-cycle,
with both haploid (homozygous) and diploid (heterozygous)
heterothallic individuals, as well as a few aneuploid/polyploid
individuals.

T.delbrueckii was formerly described as a haploid species [91—93],
because of its small cell size and also because tetrads are rarely
observed following sporulation. However, several lines of evidence
suggest that T. delbrueckii may not be haploid: first, the recent
genome sequencing of type strain CBS 1146 reveals that, at the
genetic level, T. delbrueckii possesses apparently functional mating-
type [MAT) locus and silent HMR and HML loci, suggesting this
species could be homothallic [77]. Secondly, all 110 strains we
studied here (homozygous and heterozygous) were able to
sporulate and mostly formed ascii with a single ascospore. Strains
isolated under the name of Candida colliculosa, that should
theoretically represent the anamorphic forms, displayed similar
sporulation abilities to their so-called teleomorph counterparts. In
particular, the clinical isolates showed the same sporulation
characteristics as their non-clinical counterparts and were
distributed on the dendrogram tree, indicating that T. delbrueckii
is an opportunistic pathogen rather than an actual human
pathogen. It has to be noted that all T. delbrueckii strains sporulated
on traditional sporulation (acetate) medium after 3 days, but also
after 7-20 days on YPD-agar plates, indicating that, unlike S.

cerevisiae, starvation is not necessary for sporulation [94] . Thirdly,
our genotyping data are in accordance with the hypothesis that it
is a diploid species (with the identification of 30% of heterozygous
individuals), associated with frequent inbreeding and thus frequent
diploid homozygous individuals (65%). Such high inbreeding
could be explained by the effect of homothallism on population
genetic structure, furthermore enhanced by the aptitude of T.
delbrueckii to sporulate without starvation. The occasional identi-
fication of strains with three or four alleles for a few loci suggests
either a gene duplication ability, aneuploidy or even polyploidy, as
for S. cerevisiae [5,95]. All these results suggest a life cycle identical
to the homothallic diploid S. cerevisiae yeast. Further experiments,
like micro-dissection and genetic analysis of T. delbrueckii mono-
sporic clones, construction of haploid heterothallic strains, etc., will
elucidate definitively the life-cycle of T. delbrueckii species.

Toward genetic improvement of T. delbrueckii species for
industrial purpose

Species with biotechnological interest are usually improved for
industrial purpose, through selection experiments, breeding
programs, QTL detection, etc. This is the case of the yeast
Saccharomyces cerevisiae for which several improvement programs are
running for different technological applications (winemaking,
bakery, brewery, distillery, etc.) [96,97].

Among the non-conventional yeasts naturally associated with
food processes, T delbrueckii is particularly interesting: in wine-
making, T. delbrueckii allows reducing organoleptic defects (hydro-
gen sulphide, volatile phenols, volatile acidity) [16,17] and
increases sensorial complexity [18-20]. Thus, recendy, several
strains of T. delbrueckii were commercialized for winemaking
purpose with success, to be used in association with S. cerevisiae.
Besides winemaking, microsatellite analysis provides evidence of
anthropic selection for other bioprocesses, and a possible
specialization for dairy process.

Here, microsatellites markers were developed and gave valuable
data regarding the life-cycle of the species, suggesting a mosdy
diploid homothallic life. A better understanding of life-cycle and
the availability of highly discriminant markers paves the way
toward further biotechnological improvement of T. delbrueckii
strains for winemaking and other bioprocesses purposes.

Supporting Information

Figure SI European localisation of the T. delbrueckii
strains used in this study.

(PDF)

Table SI Origin of Torulaspora spp. strains used in this
study. a ARC-INFRUITEC: Agricultural Research Council-
Institute for Deciduous Fruit, Vines and Wine; ARS Culture
Collection: Agricultural Research Service Culture Collection,
formerly NRRL (Northern Regional Research Laboratory);
AWMCC: AWRI Wine Microorganism Culture Collection;
BCCM/MUCL: Agroindustrial fungi & yeasts collection, My-
cotheque de l'Universite catholique de Louvain; CBS-KNAW:
Centraalbureau voor Schimmelcultures (CBS) Fungal Biodiversity
Centre, institute of the Royal Netherlands Academy of Arts and
Sciences (Koninklijke Nederlandse Akademie van Wetenschap-
pen); CIATEJ: Centro de Investigation y Asistencia en Tecnologia
y Diseno del Estado de Jalisco; CIRM-Levures: Centre Interna-
tional de Ressources Microbiennes -Levures, formerly CLIB:
Collection de Levures dTnteret Biotechnologique; CRB Oeno:
Centre de Ressources Biologiques (Enologie; CRPR: Centre de
Recherche Pernod-Ricard; DIL: Deutsches Institut fur Lebens-
mitteltechnik e.V.; D.i.S.V.A.: Dipartimento di Scienze della Vita

PLOS ONE | www.plosone.org

8

April 2014 | Volume 9 | Issue 4 | e94246

Microsatellite Analysis of T. delbrueckii

e delPAmbiente; DSMZ: Deutsche Sammlung von Mikroorganis-
men und Zellkulturen GmbH; IDEPA: Instituto Multidisciplinario
de Investigation y Desarrollo de la Patagonia Norte; IFI: Instituto
de Fermentaciones Industriales; IFV: Institut Francais de la Vigne
et du Vin; ISW: Institut des Sciences de la Vigne et du Vin;
IWBT: Institute for Wine Biotechnology; LYCC: Lallemand Yeast
Culture Collection; MRI: Max Rubner-Institut; NCAIM: Nation-
al Collection of Agricultural and Industrial Microorganisms;
NIAS: National Institute of Agrobiological Sciences; NPCC:
North Patagonian Culture Collection; PYCC: Portuguese yeast
Culture Collection; UOA/HCPF: Hellenic Collection of Patho-
genic Fungi, University of Athens; UWOPS: Culture collection of
the University of Western Ontario, Department of Biology
(formerly Plant Sciences). b Grant FONDEF D98I1037- Chile.
Collection and characterization of native yeast strains for
differentiation and identity of Chilean wine (1999-2003).
(XLSX)

Table S2 Genotype of 110 strains of Torulaspora
delbrueckii using eight microsatellite markers. For each
strain and each marker, the size of the amplicons is indicated. For
heterozygous loci, the different alleles are separated by a slash. NA
stands for "Not Available" (missing) data.
(XLSX)

References

1. Samuel D (1996) Investigation of ancient egyptian baking and brewing methods
by correlative microscopy. Science 273: 488—490.

2. Cavahcri D, McGovern PE, Hard DL, Mortimer R, Polsinelli M (2003)
Evidence for S. cerevisiae fermentation in ancient wine. J Mol Evol 57 Suppl 1:
S226-232.

3. Mortimer RK (2000) Evolution and variation of the yeast (Saccharomyces) genome.
Genome Res 10: 403^-09.

4. Galbe M, Zacchi G (2002) A review of the production of ethanol from softwood.
Applied Microbiology and Biotechnology 59: 618-628.

5. Legras JL, Merdinoglu D, Gornuet JM, Karst F (2007) Bread, beer and wine:
Saccharomyces cerevisiae diversity reflects human history. Mol Eeol 16: 2091-2102.

6. Douglas GL, Klaenhammer TR (2010) Genomic evolution of domesticated
microorganisms. Annu Rev Eood Sci Teehnol 1: 397—414.

7. Gellissen G (2005) Hansen ula po/ymorplia: Biology and Applications: Wilcy-VCH
Verlag GmbH & Co. KGaA. i-xiii p.

8. Kim Y-J, Oh Y-K, Kang W, Lee E, Park S (2005) Production of human
caseinomacropeptidc in recombinant Saccharomyces cerevisiae and Pichia pastoris.
Journal of Industrial Microbiology and Biotechnology 32: 402-408.

9. Bonekamp F, Oostcrom J (1994) On the safety of ' Kluyveromyces lactis — a review.
Applied Microbiology and Biotechnology 41: 1-3.

10. Castelli T (1954) Les agents de la fermentation vinaire. Archiv fur Mikrobiologie
20: 323-342.

11. van Breda V, Jolly N, van WykJ (2013) Characterisation of commercial and
natural Torulaspora delbrueckii wine yeast strains. International Journal of Food
Microbiology 163: 80-88.

12. Sangorrin M, Lopes C, Jofrc V, Querol A, Caballero A (2008) Spoilage yeasts
from Patagonian cellars: characterization and potential biocontrol based on
killer interactions. World Journal of Microbiology and Biotechnology 24: 945-
953.

13. Parle JN, Di Menna ME (1966) The source of yeasts in New Zealand wines. New
Zealand Journal of Agricultural Research 9: 98-107.

14. Castelli T (1940) Sulla validita del genere Torulaspora. Archiv fur Mikrobiologie
11.

15. Comitini F, Gobbi M, Domizio P, Romani G, Lencioni L, et al. (201 1) Selected
non- Saccharomyces wine yeasts in controlled multistarter fermentations with
Saccharomyces cerevisiae. Food Microbiol 28: 873-882.

16. Renault P, Miot-Sertier C, Marullo P, Hernandez-Ortc P, Lagarrigue L, et al.
(2009) Genetic characterization and phenotypic variability in Torulaspora
delbrueckii species: Potential applications in the wine industry. Int J Food
Microbiol 134: 201-210.

17. Bely M, Stoecklc P, Masneuf-Pomarede I, Dubourdieu D (2008) Impact of
mixed Torulaspora delbrueckii-Saecharomyces cerevisiae culture on high-sugar fermen-
tation. IntJ Food Microbiol 122: 312-320.

18. Azzolini M, Fedrizzi B, Tosi E, Finato F, Vagnoli P, et al. (2012) Effects of
Torulaspora delbrueckii and Saccharomyces cerevisiae mixed cultures on fermentation
and aroma of Amarone wine. European Food Research and Technology 235:
303-313.

19. Ciani M, Maccarclli F (1998) Oenological properties of non- Saccharomyces yeasts
associated with wine-making. World Journal of Microbiology & Biotechnology
14: 199-203.

Acknowledgments

We kindly thank the following persons for providing strains and/ or advice:
Ailsa Hocking, Marc-Andre Lachance, Andreas Podbielski, Audrey
Yurkov, Angelica Ganga, Anne Gschaedler, Benoit Colonna-Ceccaldi,
Benoit Divol, Celia Quintas, Christian Hertel, Christian von Wallbrunn,
David Santo, Dennis Sandris Nielsen, Erna Storgards, Eveline Bartowsky,
Gabor Peter, Guilherme Martins, Hagen Frickmann, Harald Claus,
Hector R Urbina, Heide-Maric Daniel, Herve Alexandre, Hiroshi
Fukuhara, Huu-Vang Nguyen, Irena Macioniene, Isabelle Masneuf-
Pomarede, James Swezey,Jane McCarthy, Juan Jose Rubio Coque, Keith
Richards, Knut J. Heller, Kyria Boundy-Mills, Lucy Joseph, Makoto
Kawase, Marcela Sangorrin, Mark Wilson, Matthias Sipiczki, Maurizio
Ciani, Melanie Wieschebrock, Meredith Blackwell, Michael Arabatzis,
Monika Coton, Morvan Coarer, Neil Jolly, Noemie Jacques, Patrizia
Romano, Petra Wrent, Purificacion Hernandez-Orte, Savitree Limtong,
Serge Casaregola, Joana Coulon and Aline Lonvaud-Funel. We thank
Caroline Ahern for correcting the manuscript.

Author Contributions

Conceived and designed the experiments: WA PM MB. Performed the
experiments: WA LC GC AP. Analyzed the data: WA LC PM MB.
Contributed reagents/materials/analysis tools: AD FS. Wrote the paper:
WA LC PM MB.

20. Herraiz T, Reglero G, Hcrraiz M, Martin-Alvarez PJ, Cabczudo MD (1990)
The Influence of the Yeast and Type of Culture on the Volatile Composition of
Wines Fermented Without Sulfur Dioxide. American Journal of Enology and
Viticulture 41: 313-318.

21. Renault PE, Albertin W, Bely M (2013) An innovative tool reveals interaction
mechanisms among yeast populations under oenological conditions. Applied
Microbiology and Biotechnology 97: 4105-4119.

22- Nisscn P, Nielsen D, Arneborg N (2003) Viable Saccharomyces cerevisiae cells at high
concentrations cause early growth arrest of non-Saccharomyces yeasts in mixed
cultures by a cell-cell contact-mediated mechanism. Yeast 20: 331-341.

23. Nisscn P, Arneborg N (2003) Characterization of early deaths of non-
Saccharomyces yeasts in mixed cultures with Saccharomyces cerevisiae. Arch Microbiol

180: 257-263.

24. Pacheeo A, SantosJ, Chaves S, AlmcidaJ, Leao C, et al. (2012) The Emerging
Role of the Yeast Torulaspora delbrueckii in Bread and Wine Production: Using
Genetic Manipulation to Study Molecular Basis of Physiological Responses. In:
Eissa AHA, editor. Structure and Function of Food Engineering: InTech.

25. Hernandez-Lopez MJ, Prieto JA, Randez-Gil F (2003) Osmotolerance and
leavening ability in sweet and frozen sweet dough. Comparative analysis
between Torulaspora delbrueckii and Saccharomyces cerevisiae baker's yeast strains.
Anionic Van Leeuwenhoek International Journal of General and Molecular
Microbiology 84: 125-134.

26. Alvcs-Araujo C, Almeida MJ, Sousa MJ, Leao C (2004) Freeze tolerance of the
yeast Torulaspora dclbrttt'ckii: cellular and biochemical basis. FEMS Microbiol Lett
240: 7-14.

27. Almeida MJ, Pais C (1996) Leavening ability and freeze tolerance of yeasts
isolated from traditional corn and rye bread doughs. Applied and Environmental
Microbiology 62: 4401-4404.

28. Vrancken G, Dc Vuyst L, Van der Meulcn R, Huys G, Vandamme P, et al.
(2010) Yeast species composition differs between artisan bakery and spontaneous
laboratory' sourdoughs. FEMS Yeast Research 10: 471-481.

29. Nielsen DS, Snitkjaer P, van den Berg F (2008) Investigating the fermentation of
cocoa by correlating Denaturing Gradient Gel Electrophoresis profiles and Near
Infrared spectra. International Journal of Food Microbiology 125: 133-140.

30. Papalexandratou Z, Falony G, Romanens E,JimenezJC, Amores F, et al. (201 1)
Species Diversity, Community Dynamics, and Metabolite Kinetics of the
Microbiota Associated with Traditional Ecuadorian Spontaneous Cocoa Bean
Fermentations. Applied and Environmental Microbiology 77: 7698—7714.

31. Kotzekidou P (1997) Identification of yeasts from black olives in rapid system
microtitrc plates. Food Microbiology 14: 609-616.

32. Etchells JL, Costilow RN, Bell TA (1952) Identification of Yeasts from
Commercial Cucumber Fermentations in Northern Brining Areas. Farlowia 4:
249-264.

33. Etchells JL, Bell TA (1950) Classification of yeasts from fermentation of
commercially brined cucumbers. Farlowia 4: 112.

34. Verdugo Valdez A, Segura Garcia L, Kirchmayr M, Ramirez Rodriguez P,
Gonzalez Esquinca A, et al. (201 1) Yeast communities associated with artisanal
mezcal fermentations from Agave salmiana. Antonie Van Leeuwenhoek 100:
497-506.

PLOS ONE | www.plosone.org

9

April 2014 | Volume 9 | Issue 4 | e94246

Microsatellite Analysis of T. delbrueckii

35. Ulloa M, Hcrrcra T (1978) Torulopsis taboadae, una nucva cspccic dc lcvadura
aislada del eolonche dc Zacatccas, Mexico. Bolctin dc la Socicdad Mcxicana dc
Micologia 12: 5-12.

36. Lachance M-A (1995) Yeast communities in a natural tequila fermentation.
Antonic Van Lccuwcnhock 68: 151-160.

37. Coton E, Goton M, Levert D, Casaregola S, Sohier D (2006) Yeast ecology in
French cider and black olive natural fermentations. International Journal of
Food Microbiology 108: 130-135.

38. Santo DE, Galcgo L, Goncalvcs T, Quintas C (2012) Yeast diversity in the
Mediterranean strawberry tree (Arbutus unedo L.) fruits' fermentations. Food
Research International 47: 45-50.

39. Pataro C, Guerra JB, Pctrillo-Pcixoto ML, Mcndonca-Haglcr LG, Linardi VR,
et al. (2000) Yeast communities and genetic polymorphism of Saccharomyces
cerevisiae strains associated with artisanal fermentation in Brazil. Journal of
Applied Microbiology 89: 24-31.

40. Stringini M, Gomitini F, Taccari M, Giani M (2008) Yeast diversity in crop-
growing environments in Cameroon. International Journal of Food Microbiol-
ogy 127: 184-189.

41. Lorctan T, Mostcrt JF, Viljoen BC (2003) Microbial flora associated with South
African household kefir. South African Journal of Science 99: 92-94.

42. Wclthagcn JJ, Viljoen BC (1998) Yeast profile in Gouda cheese during
processing and ripening. International Journal of Food Microbiology 41: 185-
194.

43. Gadaga TH, Mutukumira AN, Narvhus JA (2000) Enumeration and
identification of yeasts isolated from Zimbabwean traditional fermented milk.
International Dairy Journal 10: 459-466.

44. Andrighetto C, Psomas E, Tzanetakis N, Suzzi G, Lombardi A (2000) Randomly
amplified polymorphic DNA (RAPD) PGR for the identification of yeasts
isolated from dairy products. Letters in Applied Microbiology 30: 5-9.

45. Westall S, Filtcnborg O (1998) Spoilage yeasts of decorated soft cheese packed in
modified atmosphere. Food Microbiology 15: 243-249.

46. Ros-Chumillas M, Egea-Cortines M, Lopez-Gomez A, Weiss J (2007)
Evaluation of a rapid DNA extraction method to detect yeast cells by PGR in
orange juice. Food Control 18: 33-39.

47. Capriotti A (1957) Tomlaspora nilssoni nov. spec. Archiv fiir Mikrobiologie 28:
247-254.

48. Limtong S, Koowadjanakul N (2012) Yeasts from phylloplanc and their
capability to produce indole- 3-acetic acid. World Journal of Microbiology and
Biotechnology 28: 3323-3335.

49. Tokuoka K, Ishitani T, Goto S, Komagata K (1985) IDENTIFICATION OF
YEASTS ISOLATED FROM HIGH-SUGAR FOODS. The Journal of
General and Applied Microbiology 31: 411-427.

50. Nguyen NH, Suh S-O, Blackwcll M (2007) Five novel Candida species in insect-
associated yeast clades isolated from Neuroptera and other insects. Mycologia 99:
842-858.

51. Nguyen NH, Suh SO, Erbil CK, Blackwcll M (2006) Metschnikowia noctiluminum
sp. nov., Metschnikowia comiflorae sp. nov., and Candida chrysomelidarum sp. nov.,
isolated from green lacewings and beetles. Mycol Res 110: 346-356.

52. Kaygusuz I, Mulazimoglu L, Cerikcioglu N, Toprak A, Oktay A, et al. (2003) An
unusual native tricuspid valve endocarditis caused by Candida colliculosa. Clinical
Microbiology and Infection 9: 319—322.

53. Frickmann H, Lakncr A, Essig A, Poppert S (2012) Rapid identification of yeast
by fluorescence in situ hybridisation from broth and blood cultures. Mycoses 55:
521-531.

54. Saluja P, Yelchuri RK, Sohal SK, Bhagat G, Paramjit, ct al. (2012) Tomlaspora
indica a novel yeast species isolated from coal mine soils. Antonie Van
Lccuwenhoek 101: 733-742.

55. Limtong S, Imanishi Y, Jindamorakot S, Ninomiya S, Yongmanitchai W, et al.
(2008) Tomlaspora maleeae sp. nov., a novel ascomyectous yeast species fromjapan
and Thailand. FEMS Yeast Res 8: 337-343.

56. Wang QM, Xu J, Wang H, Li J, Bai FY (2009) Tomlaspora quercuum sp. nov. and
Candida pscudohumilis sp. nov., novel yeasts from human and forest habitats.
FEMS Yeast Res 9: 1322-1326.

57. Granchi L, Boseo M, Messini A, Vincenzini M (1999) Rapid detection and
quantification of yeast species during spontaneous wine fermentation by PGR-
RFLP analysis of the rDNA ITS region. J Appl Microbiol 87: 949-956.

58. Kurtzman CP, Robnctt CJ (1991) Phylogenetie relationships among species of
Saccharomyces , Schizpsaccharomyces^ Debaiyomyces and Schivanniomyces determined
from partial ribosomal RNA sequences. Yeast 7: 61-72.

59. Gordon JL, Byrne KP, Wolfe KH (201 1) Mechanisms of chromosome number
evolution in yeast. PLoS Genet 7: el002190.

60. Schuclke M (2000) An economic method for the fluorescent labeling of PGR
fragments. Nat Biotechnol 18: 233-234.

61. R Development Gore Team (2010) R: A language and environment for
statistical computing. Vienna, Austria: R Foundation for Statistical Computing.

62. Clark LV, Jasieniuk M (2011) POLYSAT: an R package for polyploid
microsatellite analysis. Mol Ecol Resour 11: 562—566.

63. Paradis E, Claude J, Strimmcr K (2004) APE: Analyses of Phylogenetics and
Evolution in R language. Bioinformatics 20: 289-290.

64. Jombart T, Balloux F, Dray S (2010) adephylo: new tools for investigating the
phylogenetie signal in biological traits. Bioinformatics 26: 1907—1909.

65. Chen W-C (2011) Overlapping Codon Model, Phylogenetie Clustering, and
Alternative Partial Expectation Conditional Maximization Algorithm: Iowa
State University.

66. Bruvo R, Michiels NK, D'Souza TG, Schulcnburg H (2004) A simple method
for the calculation of microsatellite genotype distances irrespective of ploidy
level. Mol Ecol 13: 2101-2106.

67. Shimodaira H (2002) An approximately unbiased test of phylogenetie tree
selection. Systematic Biology 51: 492-508.

68. Suzuki R, Shimodaira H (2006) Pvclust: an R package for assessing the
uncertainty in hierarchical clustering. Bioinformatics 22: 1540—1542.

69. PritchardJK, Stephens M, Donnelly P (2000) Inference of Population Structure
Using Multilocus Genotype Data. Genetics 155: 945-959.

70. Paradis E (20 1 0; pegas: an R package for population genetics with an intcgratcd-
modular approach. Bioinformatics 26: 419—420.

71. Weir BS, Cockerham CC (1984) Estimating F-Statistics for the Analysis of
Population-Structure. Evolution 38: 1358-1370.

72. Goldstein DB, Ruiz Linares A, Gavalli-Sforza LL, Feldman MW (1995) Genetic
absolute dating based on microsatellites and the origin of modern humans. Proc
Natl Acad Sci U S A 92: 6723-6727.

73. Lynch M, Sung W, Morris K, Coffey N, Landry CR, et al. (2008) A genome-
wide view of the spectrum of spontaneous mutations in yeast. Proc Natl Acad
Sci USA 105: 9272-9277.

74. Pclletier R, Krasilnikova MM, Samadashwily GM, Lahue R, Mirkin SM (2003)
Replication and Expansion of Trinucleotide Repeats in Yeast. Molecular and
Cellular Biology 23: 1349-1357.

75. FayJC, Benavidcs JA (2005) Evidence for domesticated and wild populations of
Saccharomyces cerevisiae. PLoS Genet 1: 66-71.

76. Kuehne HA, Murphy HA, Francis Chantal A, Sniegowski PD (2007) Allopatric
Divergence, Secondary Contact, and Genetic Isolation in Wild Yeast
Populations. Current Biology 17: 407-411.

77. Gordon JL, Armiscn D, Proux-Wcra E, OhEigeartaigh SS, Byrne KP, et al.
(201 1) Evolutionary erosion of yeast sex chromosomes by mating-typc switching
accidents. Proceedings of the National Academy of Sciences 108: 20024—20029.

78. Valdes AM, Slatkin M, Frcimer NB (1993) Allele Frequencies at Microsatellite
Loci - the Stepwise Mutation Model Revisited. Genetics 133: 737—749.

79. Primmer GR, Ellegrcn H (1998) Patterns of molecular evolution in avian
microsatellites. Molecular Biology and Evolution 15: 997—1008.

80. Kruglyak S, Durrett RT, Schug MD, Aquadro CF (1998) Equilibrium
distributions of microsatellite repeat length resulting from a balance between
slippage events and point mutations. Proc Natl Acad SciU S A 95: 10774-10778.

81. Zott K, Claisse O, Lucas P, Goulon J, Lonvaud-Funcl A, et al. (2010)
Characterization of the yeast ecosystem in grape must and wine using real-time
PGR. Food Microbiol 27: 559-567.

82. Rosenberg NA, PritchardJK, Weber JL, Cann HM, Kidd KK, et al. (2002)
Genetic structure of human populations. Science 298: 2381—2385.

83. Schlotterer C (2001) Genealogical inference of closely related species based on
microsatellites. Genet Res 78: 209-212.

84. Schullcr D, Casal M (2007) The genetic structure of fermentative vineyard-
associated Saccharomyces cerevisiae populations revealed by microsatellite analysis.
Antonie Van Leeuwenhoek 91: 137-150.

85. Richards K, Goddard M, Gardner R (2009) A database of microsatellite
genotypes for Saccharomyces cerevisiae. Antonie Van Lccuwenhoek 96: 355-359.

86. Goddard MR, Anfang N, Tang R, Gardner RC, Jun C (2010) A distinct
population of Saccharomyces cerevisiae in New Zealand: evidence for local dispersal
by insects and human-aided global dispersal in oak barrels. Environmental
Microbiology 12: 63-73.

87. This P, Lacombc T, Thomas MR (2006) Historical origins and genetic diversity
of wine grapes. Trends in Genetics 22: 511—519.

88. Sicard D, Legras JL (2011) Bread, beer and wine: Yeast domestication in the
Saccharomyces sensu stricto complex. C R Biol 334: 229-236.

89. Stefanini I, Dapporto L, Legras JL, Calabretta A, Di Paola M, et al. (2012) Role
of social wasps in Saccharomyces cerevisiae ecology and evolution. Proc Natl Acad
Sci USA 109: 13398-13403.

90. Paolocci F, Rubini A, Riccioni C, Arcioni S (2006) Reevaluation of the Life Cycle
of Tuber magnatum. Applied and Environmental Microbiology 72: 2390-2393.

91. Hernandez-Lopez MJ, Pallotti C, Andreu P, AguileraJ, Pricto JA, et al. (2007)
Characterization of a Tomlaspora delbrueckii diploid strain with optimized
performance in sweet and frozen sweet dough. International Journal of Food
Microbiology 116: 103-110.

92. Sasaki T, Ohshima Y (1987) Induction and Characterization of Artificial
Diploids from the Haploid Yeast Tomlaspora delbrueckii. Applied and Environ-
mental Microbiology 53: 1504-1511.

93. Kurtzman CP, FcllJW, Bockhout T (2011) The Yeasts: A Taxonomic Study.
Amsterdam: Elsevier.

94. Codon AC, Gasentramirez JM, Benitez T (1995) Factors Which Affect the
Frequency of Sporulation and Tetrad Formation in Saccharomyces cerevisiae Bakers
Yeasts. Applied and Environmental Microbiology 61: 1677—1677.

95. Albertin W, Marullo P (2012) Polyploidy in fungi: evolution after wholc-gcnomc
duplication. Proceedings of the Royal Society B: Biological Sciences 279: 2497-2509.

96. Marullo P, Bely M, Masneuf-Pomarede I, Pons M, Aigle M, ct al. (2006)
Breeding strategies for combining fermentative qualities and reducing off-flavor
production in a wine yeast model. FEMS Yeast Res 6: 268-279.

97. Marullo P, Mansour G, Dufour M, Albertin W, Sicard D, et al. (2009) Genetic
improvement of thermo-tolcrance in wine Saccharomyces cerevisiae strains by a
backcross approach. FEMS Yeast Res 9: 1148-1160.

PLOS ONE | www.plosone.org

10

April 2014 | Volume 9 | Issue 4 | e94246